Patient survey.

The principal research question is: Are preoperative VLEDs with Optifast® 900 (intervention) safe, well adhered to, and well tolerated (outcomes) in adult patients with obesity (i.e., older than 18 years of age and BMI greater than 30 kg/m²) undergoing elective non-bariatric intra-abdominal surgery (population)?

The specific aims of this patient survey will be: (1) Determine patient-reported adverse effects of participating in preoperative protocols that include preoperative VLEDs with liquid formulation (Safety); (2) Determine patient-reported adherence with preoperative VLEDs with liquid formulation prior to elective non-bariatric abdominal surgery (Adherence); (3) Determine barriers to adherence with preoperative VLEDs with liquid formulation (Adherence); (4) Determine thresholds for outcome differences at which patients would be willing to adhere with preoperative VLEDs with liquid formulation (Adherence); (5) Determine the amount of patient reported preoperative weight loss with the preoperative use of VLEDs (Efficacy); (6) Determine the impact of preoperative VLEDs with liquid formulation on health-related QoL in adult patients with obesity undergoing non-bariatric abdominal surgery (Quality of Life [QoL]); and (7) Inform the design of a multi-center RCT assessing the use of preoperative VLED (Feasibility).

Provider survey.

The principal research question is: Are Canadian surgeons who perform elective non-bariatric intra-abdominal surgery (population) willing to prescribe preoperative weight loss by way of VLEDs (outcome) for patients with obesity undergoing non-bariatric abdominal surgery according to self-reported survey data (intervention)?

The specific aims of this provider survey will be: (1) Determine willingness to prescribe preoperative VLED to patients with both benign and malignant diseases undergoing non-bariatric abdominal surgery (Feasibility); (2) Determine the frequency at which preoperative weight loss interventions are prescribed for patients with obesity undergoing non-bariatric abdominal surgery (Feasibility); (3) Determine the different types of preoperative weight loss interventions that are being used in Canada (Feasibility); and (4) Inform the design of a multi-center RCT assessing the use of preoperative VLED (Feasibility).

Patient survey.

This is a cross-sectional, single-center survey.

Provider survey.

This is a cross-sectional national survey.

Patient survey.

The population of interest is all adult (i.e., 18 years of age or older) patients with obesity (i.e., BMI greater than 30kg/m²) undergoing elective non-bariatric intra-abdominal surgery at St. Joseph’s Healthcare Hamilton who were prescribed a course of preoperative VLED. If the first postoperative visit is within two months of their operative date, patients will be considered for inclusion. Patients who underwent bariatric surgery and surgeries other than intra-abdominal surgery will be excluded. Patients who are unable to either read, write, or speak in English or have any comorbidities, impairments, or disabilities prohibiting them from completing the survey will be excluded.

Provider survey.

The population of interest is all independent practicing surgeons in Canada who perform major non-bariatric abdominal surgery. Surgical sub-specialties that will be eligible for inclusion will be general surgery, colorectal surgery, hepatobiliary surgery, surgical oncology, thoracic surgery, vascular surgery, urology, and gynecology. Surgeons who perform bariatric surgery exclusively will be excluded. Surgeons who are unable to respond to the survey in English or French will be excluded. Incomplete survey responses will be excluded.

Patient survey.

Our sampling strategy will be non-probability convenience sampling in which we will approach consecutive patients presenting to our surgical clinics at St. Joseph’s Healthcare Hamilton for their first postoperative visit. Non-probability sampling refers to a non-random sample selection, of which convenience sampling is one particular sub-type. Surgeons will introduce the research study to patients presenting to their initial postoperative clinic visit. If the patient is agreeable to learning more about the study, the research team will be notified. They will confirm eligibility and then, should the patient be eligible, proceed with the informed consent process. Immediately following this, the researcher will implement the survey. The survey will be kept as a hard-copy paper survey as well as converted to an electronic form on RedCap®. For patients presenting to their first postoperative visit (i.e., within two months of their operative date) in person, we will distribute a hard-copy paper survey to be completed in clinic. For patients with a virtual postoperative visit, we will obtain their email address and distribute an electronic cope of the survey to be completed on RedCap® at their earliest convenience.

Provider survey.

The sampling strategy for this survey will also be non-probability convenience sampling. To distribute the survey as broadly as possible, we will distribute the electronic survey via email through provincial and national surgery associations (e.g., Canadian Association of General Surgeons, Canadian Urology Association, Society of Obstetricians and Gynecologists, Ontario Association of General Surgeons, Association Québécoise de Chirurgie, etc.) listservs. If the response rate remains low (i.e., sample size goal not reached within 3 months) after distribution through these channels, we will contact individual departments of surgery at each academic institution across Canada (i.e., University of British Columbia, University of Calgary, University of Alberta, University of Saskatchewan, University of Manitoba, Northern Ontario School of Medicine, University of Western Ontario, McMaster University, University of Toronto, Queen’s University, University of Ottawa, McGill University, University of Sherbrooke, University of Montreal, Laval University, Dalhousie University, Memorial University) as well as the corresponding divisions of General Surgery to distribute to their members.

Survey design.

The surveys will be designed according to the Canadian Medical Association Journal Guide for the Design and Conduct of Self-Administration Surveys for Clinicians [13]. A thorough literature review informed survey questions. Content experts (one colorectal surgeon, one general surgeon, and two bariatric surgeons) collaborated to develop survey questions. Prior to production of the final version of the surveys, methodology was critically appraised by two PhD biostatisticians. The patient survey was piloted by five bariatric surgery patients who had completed a preoperative VLED prior to their operation. The provider survey was piloted by five local practicing surgeons prior to dissemination to the sample population. Previous survey methodology research has demonstrated the importance of piloting surveys in an attempt to reduce bias and improve accuracy [14]. Pilot responses will not be included in the final analyses. The provider survey was translated to French by a member of the study team who is a practicing General Surgeon in Montreal and is bilingual.

Patient survey.

The patient survey will collect data as per the following survey sections: (1) Demographic information (e.g., age, sex, location of residence); (2) Treatment information (e.g., date of surgery, type of disease, type of surgery, operative approach); (3) Prescribed preoperative weight loss information (e.g., duration of VLED, adherence with VLED, barriers to adherence, VLED-associated adverse events); and (4) Health-related QoL (i.e., according to the Short-Form 12).

Provider survey.

The provider survey will collect data as per the following survey sections: (1) Demographic information (e.g., age, sex, number of years in practice, location of practice, type of practice); (2) Institution information (e.g., number of years in practice, location of practice, type of practice, availability of bariatric surgery, availability of dietician, availability of non-surgical preoperative clinics); (3) Preoperative weight loss prescribing practices, willingness, and knowledge (e.g., is prescribing preoperative weight loss a routine part of practice, what types of preoperative weight loss interventions are prescribed, what BMI cut-off is utilized); (4) Questions specific to patients with oncologic disease (e.g., apprehension with regards to prescribing to patients with oncologic disease, willingness to prescribing to patients with oncologic disease).

Patient survey.

The primary outcomes will be safety and adherence. Safety will be assessed through VLED-associated adverse events. The adverse events will be recorded as dichotomous outcomes and described as either minor or serious, in a similar fashion to the OPTIWIN Study; the largest medical weight loss RCT evaluating Optifast® [15]. Minor adverse events will include constipation, diarrhea, nausea, fatigue, dizziness, headache, and alopecia. Serious adverse events will include acute kidney injury, symptomatic electrolyte disturbance, cardiac arrhythmias, symptomatic cholelithiasis, pancreatitis, pyelonephritis, and gout. Preoperative VLED adherence will be calculated as the number of doses of preoperative VLED liquid formulation taken as a proportion of the total doses prescribed as reported by the patient to the best of their abilities.

Secondary outcomes will include: (1) Barriers to preoperative VLED adherence; (2) Thresholds for outcome differences at which patients would be willing to adhere with preoperative VLEDs with liquid formulation; (3) Preoperative weight loss. This will be calculated as the patient’s weight prior to VLED implementation in kilograms subtracted by the patient’s weight immediately preoperatively in kilograms. Height in centimeters will also be calculated and BMI will be computed. We will also compute percent total body weight loss as a measure of weight loss that controls for baseline weight; and (4) Health-related QoL. This will be assessed with an adapted version of the Short-Form 12 (SF-12). This is a truncated version of the SF-36 and has been validated in patients with cardiovascular disease and chronic kidney disease [16,17].

Provider survey.

The primary outcome will be willingness to prescribe preoperative VLED to patients with obesity undergoing major non-bariatric abdominal surgery for both benign and malignant indications. This will be assessed through five-point Likert scale responses. Secondary outcomes will include: (1) Frequency and type(s) of preoperative weight loss interventions currently being prescribed by practicing surgeons in Canada for patients with obesity undergoing non-bariatric abdominal surgery; (2) Barriers to prescribing preoperative weight loss interventions; (3) Factors associated with prescribing preoperative weight loss; (4) Perceived benefits of prescribing preoperative weight loss; (5) Knowledge surrounding preoperative weight loss options; (6) Perceived difficulty of operating on patients with obesity for major abdominal surgery. These outcomes will be assessed through a combination of five-point Likert scale responses as well as narrative responses provided by surgeons.

Patient survey.

We propose 35 survey responses to adequately assess our primary outcomes. We justified out proposed sample size based on 95% confidence intervals (CIs) for our safety outcome. Specifically, we hypothesize that 32.5% of patients will experience VLED-associated adverse events based on our previous systematic review and meta-analysis evaluating adult patients with obesity receiving preoperative VLEDs prior to undergoing non-bariatric surgery. Based on review with local surgeons, they would be willing to accept a 50% risk of adverse events for their patients with the use of preoperative VLEDs, as long as the majority of these adverse events were minor (as classified above). Therefore, we calculated the number of patients that would be required to provide sufficiently narrow 95% CIs that would exclude a proportion of 50% assuming that 32.5% of patients reported adverse events. The cii proportions function in STATA version 18 was utilized and it was calculated that a sample size of 35 would be required to produce confidence intervals ranging from 16.9% to 49.2%, which we believed was adequate precision. This threshold will be assessed in the present survey to ensure these assumptions are congruent with patient thoughts and expectations.

The cohort of surgeons prescribing preoperative VLEDs for elective non-bariatric intra-abdominal surgery perform between two and six elective intra-abdominal operations per week. Approximately 40% of the patients undergoing intra-abdominal colorectal surgery at our center are living with obesity [7]. Therefore, we expect that one-to-two patients per week will be eligible for inclusion in this survey study. If the survey period lasts for six months, as planned, then we would expect that 24–48 patient survey responses could be collected. Thus, our achieving our planned sample size within our planned study timeline is reasonable.

Provider survey.

The sample size was calculated using methodology for determining survey sample size with a Likert scale primary outcome published by Park & Jung [18]. In this study, the primary Likert scale outcome that was utilized to calculate the sample size pertained to willingness to prescribe preoperative VLEDs prior to non-bariatric abdominal surgery. The following formula was utilized:

The following are the definitions of the above variables: n, sample size estimate; z, z-score; α, accepted type I error; k, number of items on the Likert scale; D, relative tolerable error in responses; C, coefficient of variation of the population; p, pairwise correlation coefficient. The z-score associated with the accepted type I error is 1.96. The relative tolerable error was set at 5%. The coefficient of variation was set at 0.3 (i.e., the standard deviation of responses will be half the value of the mean) given that respondents tend to avoid extreme responses in Likert scales. The pairwise correlation coefficient was set at 0.3 as the population is relatively heterogenous (i.e., surgeons practicing a variety of different surgical sub-specialties in a variety of different settings across Canada). Given these assumptions and using sample size tables provided by Park & Jung, the required sample size is 60.85, which was rounded to 61 [18].

According to 2020 data from the Canadian Institute for Health Information, there are 2,105 practicing general surgeons in Canada. It is estimated that 115 surgeons in Canada perform exclusively bariatric surgery [19]. If the response rate is estimated conservatively at 10%, this would result in a sample population of 200 surgeons. Therefore, recruitment to a planned sample size of 61 should be feasible within the planned survey timeline of six months.